534 Allen, Evolution of Bird-song. [o" t k 



the development of bird-song might be stated as follows: Let S 

 represent the song first developed out of the call-notes of a certain 

 species. S becomes elaborated as SS under stress of unusual 

 emotion, and SS becoming fixed in the psychology of the species, 

 the bird has two songs, S and SS, the latter a special mating-song 

 uttered only at times of great sexual excitement; then SS tends to 

 become the ordinary song, and a further elaboration, SSS, is evolved 

 to express the unusual emotion for which SS is no longer adequate. 



This process may go on indefinitely but so slowly that only in 

 rare instances can we see any evidence of it. Do we not get a 

 glimpse of it, however, in the case of the Baltimore Oriole (Icterus 

 galbula)'! Besides the harsh, chattering call which is suggestive of 

 family relationships, this bird utters clear, pleasing whistles which 

 are evidently in the nature of song-notes without amounting to 

 actual songs. Out of these separate song-notes (S) has developed 

 apparently the characteristic "ordinary song" of the Oriole (SS); 1 

 and out of this in turn has come the longer and more beautiful 

 mating-song (SSS) which is so often uttered on the wing. (This 

 is, of course, only an outline sketch of the possible development of 

 these songs, and I do not mean to imply that there were no inter- 

 mediate stages.) Here we seem to have three stages in song- 

 development still in existence. It is conceivable that a fourth 

 may be added in future ages and that the first or the second may 

 eventually be dropped from the Oriole's repertoire. 



Having elaborated this theory at some length, I have to confess 

 that it remains only a theory, and I ought, perhaps, to apologize 

 for presenting it in its present "half-baked" condition. If my 

 presentation of it, however, leads to the presentation of further 

 evidence or argument in favor of Mr. Saunders's view, or if some 

 one can show that "ordinary" songs and "mating" songs origi- 

 nated quite independently of each other, I shall be satisfied. One 

 objection that may be raised to the theory of progressive improve- 

 ment from S to SS, etc., is, of course, the marked differences 



1 I think I am justified in speaking of the Baltimore Oriole's "ordinary song," for though 

 the song is subject to so much individual variation that hardly any two birds sing the 

 same tune, yet its quality is highly characteristic; it is never to be mistaken for the song of 

 any other species with which the Oriole is commonly found, and in that sense it is a very 

 definite entity. 



